公开(公告)号：US20150022537A1

公开(公告)日：2015-01-22

申请号：US13946977

申请日：2013-07-19

Applicant: NVIDIA Corporation

Inventor： Eric B. Lum ,  Rouslan L. Dimitrov ,  Ignacio Llamas ,  Patrick James Neill ,  Yury Uralsky ,  Albert Meixner

IPC: G06T1/60 ,  G06T1/00

CPC classification number: G06T11/40

Abstract: A system, method, and computer program product are provided for shading primitive fragments. A target buffer may be recast when shaded samples that are covered by a primitive fragment are generated at a first shading rate using a first sampling mode, the shaded samples are stored in the target buffer that is associated with the first sampling mode and the first shading rate, a second sampling mode is determined, and the target buffer is associated with the second sampling mode. A sampling mode and/or shading rate may be changed for a primitive. A primitive fragment that is associated with a first sampling mode and a first shading rate is received and a second sampling mode is determined for the primitive fragment. Shaded samples corresponding to the primitive fragment are generated, at a second shading rate, using the second sampling mode and the shaded samples are stored in a target buffer.

Abstract translation: 提供了一种系统，方法和计算机程序产品，用于着色原始片段。 当使用第一采样模式以第一阴影率产生由原始片段覆盖的阴影样本时，可以重写目标缓冲器，阴影样本存储在与第一采样模式和第一着色相关联的目标缓冲器中 速率，确定第二采样模式，并且目标缓冲器与第二采样模式相关联。 对于原语，可以改变采样模式和/或阴影率。 接收与第一采样模式和第一遮蔽速率相关联的原始片段，并且为原始片段确定第二采样模式。 使用第二采样模式以第二遮蔽速率生成与原始片段相对应的阴影样本，并将阴影样本存储在目标缓冲器中。

公开(公告)号：US20140176578A1

公开(公告)日：2014-06-26

申请号：US13722901

申请日：2012-12-20

Applicant: NVIDIA CORPORATION

Inventor： Albert Meixner

IPC: G06T1/20

CPC classification number: G06T1/20 ,  G06T15/20 ,  G06T2210/52

Abstract: An input output connector for a graphics processing unit having a graphics pipeline including fixed function units and programmable function units is disclosed. Additionally, a graphics processing unit and a method of operating a graphics pipeline are disclosed. In one embodiment, the input output connector includes: (1) a request arbiter configured to connect to each of the programmable function units, receive fixed function requests therefrom and arbitrate the requests and (2) fixed unit converters, wherein each of the fixed unit converters is dedicated to a single one of the fixed function units and is configured to convert the requests directed to the single one to an input format for the single one.

Abstract translation: 公开了具有包括固定功能单元和可编程功能单元的图形管线的图形处理单元的输入输出连接器。 此外，公开了图形处理单元和操作图形流水线的方法。 在一个实施例中，输入输出连接器包括：（1）被配置为连接到每个可编程功能单元的请求仲裁器，从其接收固定功能请求并仲裁请求;以及（2）固定单元转换器，其中固定单元 转换器专用于单个固定功能单元，并且被配置为将针对单个功能单元的请求转换为单个单元的输入格式。

公开(公告)号：US20140173611A1

公开(公告)日：2014-06-19

申请号：US13713305

申请日：2012-12-13

Applicant: NVIDIA CORPORATION

Inventor： Albert Meixner

IPC: G06F9/50

CPC classification number: G06F9/50 ,  G06F9/3851 ,  G06F9/4843 ,  G06F2209/5018

Abstract: A system and method for launching data parallel and task parallel application threads. In one embodiment, the system includes: (1) a global thread launcher operable to retrieve a launch request from a queue and track buffer resources associated with the launch request and allocate output buffers therefor and (2) a local thread launcher associated with a streaming multiprocessor and operable to receive the launch request from the global thread launcher, set a program counter and resource pointers of pipelines of the streaming multiprocessor and receive reports from pipelines thereof as threads complete execution.

Abstract translation: 一种用于启动数据并行和任务并行应用程序线程的系统和方法。 在一个实施例中，系统包括：（1）全局线程启动器，其可操作以从队列检索启动请求并跟踪与启动请求相关联的缓冲器资源并为其分配输出缓冲器，以及（2）与流式传输相关联的本地线程启动器 多处理器并且可操作以从全局线程启动器接收启动请求，设置流式多处理器的管线的程序计数器和资源指针，并且从线程的线程接收报告作为线程完成执行。

公开(公告)号：US09286114B2

公开(公告)日：2016-03-15

申请号：US13713305

申请日：2012-12-13

Applicant: Nvidia Corporation

Inventor： Albert Meixner

IPC: G06F9/48 ,  G06F9/50 ,  G06F9/46 ,  G06F9/38

CPC classification number: G06F9/50 ,  G06F9/3851 ,  G06F9/4843 ,  G06F2209/5018

Abstract: A system and method for launching data parallel and task parallel application threads. In one embodiment, the system includes: (1) a global thread launcher operable to retrieve a launch request from a queue and track buffer resources associated with the launch request and allocate output buffers therefor and (2) a local thread launcher associated with a streaming multiprocessor and operable to receive the launch request from the global thread launcher, set a program counter and resource pointers of pipelines of the streaming multiprocessor and receive reports from pipelines thereof as threads complete execution.

Abstract translation: 一种用于启动数据并行和任务并行应用程序线程的系统和方法。 在一个实施例中，系统包括：（1）全局线程启动器，其可操作以从队列检索启动请求并跟踪与启动请求相关联的缓冲器资源并为其分配输出缓冲器，以及（2）与流式传输相关联的本地线程启动器 多处理器并且可操作以从全局线程启动器接收启动请求，设置流式多处理器的管线的程序计数器和资源指针，并且从线程的线程接收作为线程完成执行的报告。

公开(公告)号：US20140225902A1

公开(公告)日：2014-08-14

申请号：US13764416

申请日：2013-02-11

Applicant: NVIDIA CORPORATION

Inventor： Qiuling Zhu ,  Navjot Garg ,  Yun-Ta Tsai ,  Kair Pulli ,  Albert Meixner

IPC: G06T1/00

CPC classification number: G06T1/60 ,  G06T1/20

Abstract: An image pyramid processor and a method of multi-resolution image processing. One embodiment of the image pyramid processor includes: (1) a level multiplexer configured to employ a single processing element to process multiple levels of an image pyramid in a single work unit, and (2) a buffer pyramid having memory allocable to store respective intermediate results of the single work unit.

Abstract translation: 一种图像金字塔处理器和多分辨率图像处理方法。 图像金字塔处理器的一个实施例包括：（1）电平多路复用器，其被配置为使用单个处理元件来处理单个工作单元中的图像金字塔的多个级别，以及（2）具有可分配以存储各个中间件的存储器的缓冲金字塔 单个工作单位的结果。

公开(公告)号：US09355483B2

公开(公告)日：2016-05-31

申请号：US13946977

申请日：2013-07-19

Applicant: NVIDIA Corporation

Inventor： Eric B. Lum ,  Rouslan L. Dimitrov ,  Ignacio Llamas Ubieto ,  Patrick James Neill ,  Yury Uralsky ,  Albert Meixner

IPC: G06T11/40

CPC classification number: G06T11/40

Abstract: A system, method, and computer program product are provided for shading primitive fragments. A target buffer may be recast when shaded samples that are covered by a primitive fragment are generated at a first shading rate using a first sampling mode, the shaded samples are stored in the target buffer that is associated with the first sampling mode and the first shading rate, a second sampling mode is determined, and the target buffer is associated with the second sampling mode. A sampling mode and/or shading rate may be changed for a primitive. A primitive fragment that is associated with a first sampling mode and a first shading rate is received and a second sampling mode is determined for the primitive fragment. Shaded samples corresponding to the primitive fragment are generated, at a second shading rate, using the second sampling mode and the shaded samples are stored in a target buffer.

Abstract translation: 提供了一种系统，方法和计算机程序产品，用于着色原始片段。 当使用第一采样模式以第一阴影率产生由原始片段覆盖的阴影样本时，可以重写目标缓冲器，阴影样本存储在与第一采样模式和第一着色相关联的目标缓冲器中 速率，确定第二采样模式，并且目标缓冲器与第二采样模式相关联。 对于原语，可以改变采样模式和/或阴影率。 接收与第一采样模式和第一遮蔽速率相关联的原始片段，并且为原始片段确定第二采样模式。 使用第二采样模式以第二遮蔽速率生成与原始片段对应的阴影样本，并将阴影样本存储在目标缓冲器中。

公开(公告)号：US20140176577A1

公开(公告)日：2014-06-26

申请号：US13722771

申请日：2012-12-20

Applicant: NVIDIA CORPORATION

Inventor： Albert Meixner

IPC: G06T1/20

CPC classification number: G06T1/20 ,  G06F8/456 ,  G06F2201/865

Abstract: A method of operating a graphics pipeline, a graphics processing unit and a GPU computing system are provided by this disclosure. In one embodiment, the graphics processing unit includes: (1) a processor configured to assist in operating the graphics processing unit and (2) a graphics pipeline coupled to the processor and including a programmable shader stage, the programmable shader stage configured to determine occurrence of a pipeline exception during execution of the graphics pipeline, initiate preempting the execution in response to determining the occurrence and initiate resolving the pipeline exception before execution is restarted.

Abstract translation: 本公开提供了一种操作图形流水线，图形处理单元和GPU计算系统的方法。 在一个实施例中，图形处理单元包括：（1）处理器，被配置为帮助操作图形处理单元和（2）耦合到处理器并包括可编程着色器级的图形管线，该可编程着色器级被配置为确定发生 在执行图形流水线期间发生流水线异常，响应于确定发生而启动抢占执行，并且在执行重新启动之前启动解决流水线异常。

公开(公告)号：US09123128B2

公开(公告)日：2015-09-01

申请号：US13724233

申请日：2012-12-21

Applicant: Nvidia Corporation

Inventor： Albert Meixner

IPC: G06T1/20 ,  H05K13/00 ,  G06F15/76 ,  G06F9/38

CPC classification number: G06T1/20 ,  G06F9/3887 ,  G06F15/76 ,  G06F15/7867 ,  G06F15/8053 ,  H05K13/00 ,  Y10T29/49

Abstract: Employing a general processing unit as a programmable function unit of a graphics pipeline and a method of manufacturing a graphics processing unit are disclosed. In one embodiment, the graphics pipeline includes: (1) accelerators, (2) an input output interface coupled to each of the accelerators and (3) a general processing unit coupled to the input output interface and configured as a programmable function unit of the graphics pipeline, the general processing unit configured to issue vector instructions via the input output interface to vector data paths for the programmable function unit.

Abstract translation: 使用通用处理单元作为图形管线的可编程功能单元和制造图形处理单元的方法。 在一个实施例中，图形管线包括：（1）加速器，（2）耦合到每个加速器的输入输出接口和（3）耦合到输入输出接口并被配置为可编程功能单元的通用处理单元 图形管线，通用处理单元被配置为通过输入输出接口向可编程功能单元的矢量数据路径发出矢量指令。

公开(公告)号：US09019284B2

公开(公告)日：2015-04-28

申请号：US13722901

申请日：2012-12-20

Applicant: Nvidia Corporation

Inventor： Albert Meixner

IPC: G06T1/00 ,  G06T1/20

CPC classification number: G06T1/20 ,  G06T15/20 ,  G06T2210/52

Abstract: An input output connector for a graphics processing unit having a graphics pipeline including fixed function units and programmable function units is disclosed. Additionally, a graphics processing unit and a method of operating a graphics pipeline are disclosed. In one embodiment, the input output connector includes: (1) a request arbiter configured to connect to each of the programmable function units, receive fixed function requests therefrom and arbitrate the requests and (2) fixed unit converters, wherein each of the fixed unit converters is dedicated to a single one of the fixed function units and is configured to convert the requests directed to the single one to an input format for the single one.

Abstract translation: 公开了具有包括固定功能单元和可编程功能单元的图形管线的图形处理单元的输入输出连接器。 此外，公开了图形处理单元和操作图形流水线的方法。 在一个实施例中，输入输出连接器包括：（1）被配置为连接到每个可编程功能单元的请求仲裁器，从其接收固定功能请求并仲裁请求;以及（2）固定单元转换器，其中固定单元 转换器专用于单个固定功能单元，并且被配置为将针对单个功能单元的请求转换为单个单元的输入格式。

公开(公告)号：US20140176569A1

公开(公告)日：2014-06-26

申请号：US13724233

申请日：2012-12-21

Applicant: NVIDIA CORPORATION

Inventor： Albert Meixner

IPC: G06T1/20 ,  H05K13/00

CPC classification number: G06T1/20 ,  G06F9/3887 ,  G06F15/76 ,  G06F15/7867 ,  G06F15/8053 ,  H05K13/00 ,  Y10T29/49

Abstract: Employing a general processing unit as a programmable function unit of a graphics pipeline and a method of manufacturing a graphics processing unit are disclosed. In one embodiment, the graphics pipeline includes: (1) accelerators, (2) an input output interface coupled to each of the accelerators and (3) a general processing unit coupled to the input output interface and configured as a programmable function unit of the graphics pipeline, the general processing unit configured to issue vector instructions via the input output interface to vector data paths for the programmable function unit.

Abstract translation: 使用通用处理单元作为图形管线的可编程功能单元和制造图形处理单元的方法。 在一个实施例中，图形管线包括：（1）加速器，（2）耦合到每个加速器的输入输出接口和（3）耦合到输入输出接口并被配置为可编程功能单元的通用处理单元 图形管线，通用处理单元被配置为通过输入输出接口向可编程功能单元的矢量数据路径发出矢量指令。